On Sun, 2007-11-04 at 10:54 +0000, omat wrote:
> I change one thing at a time when debugging but after posting here, I
> go on with experimenting. And when I respond to a request, like
> posting the traceback, usually it is not the very exact instance that
> I am running currently.
Well, there's a problem right there. If the code you are posting here is
not the same code you are using to create the traceback, there are going
to be (possibly important differences). I now have little confidence
that what you're showing us matches the real problem. Construct the
smallest example possible that shows the issue so that people can repeat
it. That's the usual procedure. You said your data had type unicode, but
it looks like (below) it's actually a bytestring, for example. The
differences are very important in cases like this.
You might want to try subversion trunk, since I added extra debugging to
smart_unicode and force_unicode so that they show the data you pass in
and it's type as part of the output. Maybe that will give you more
information.
> I change it to be consistent with the on going
> discussion, but sometimes there are some portions that are not
> matched. But they have the same nature of problem.
>
> In our case, the message and the subject have the same problem and
> generated almost identical tracebacks, when instance.name contains non-
> ascii chars.
>
> And here is my Tag model:
>
> class Tag(models.Model):
> name = models.CharField(max_length=50,
> db_index=True,
> validator_list=[isTag])
> slug = models.SlugField(editable = False,
> unique = True)
> ignore = models.BooleanField(default=False)
> synonym_for = models.ForeignKey('self', blank=True, null=True)
>
> objects = TagManager()
>
> def save(self):
> self.slug = slugify(self.name)
> super(Tag, self).save()
>
> def __unicode__(self):
> name = self.name
> if self.ignore:
> name = u'%s (x)' % (name)
> if self.synonym_for:
> name = u'%s => %s' % (name, self.synonym_for.name)
> return name
>
I don't see how you're getting this problem with self.name being a
Unicode object, so let's confirm that. In fact, it looks a lot like
self.name would be a bytestring -- that would certainly explain the
UnicodeDecodeError, because you cannot substitute non-ASCII bytestrings
into a u'%s' format string.
If 'name' is a bytestring (or if self.synonym_for.name is a bytestring),
you'll need to manually call force_unicode() on them before trying to
substitute them into the u'%s => %s' bit.
Anyway, good luck with your problem, but until the code you are showing
can be used to repeat the problem and isn't a variation that you have
changed without knowing what is significant, I cannot help any further.
Malcolm
--
If at first you don't succeed, destroy all evidence that you tried.
http://www.pointy-stick.com/blog/
--~--~---------~--~----~------------~-------~--~----~
You received this message because you are subscribed to the Google Groups
"Django users" group.
To post to this group, send email to django-users@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at
http://groups.google.com/group/django-users?hl=en
-~----------~----~----~----~------~----~------~--~---
